REVIEW 7 cited by
FalseReject: A Resource for Improving Contextual Safety and Mitigating Over-Refusals in LLMs via Structured Reasoning
Not yet reviewed by Pith; the record is open.
This paper has not been read by Pith yet. Machine review is queued; the pith claim, tier, and objections will appear here once it completes.
SPECIMEN: schema-true, not a live event
T0 review · schema-true
One-sentence machine reading of the paper's core claim.
pith:XXXXXXXX · record.json · timestamp
read the original abstract
Safety alignment approaches in large language models (LLMs) often lead to the over-refusal of benign queries, significantly diminishing their utility in sensitive scenarios. To address this challenge, we introduce FalseReject, a comprehensive resource containing 16k seemingly toxic queries accompanied by structured responses across 44 safety-related categories. We propose a graph-informed adversarial multi-agent interaction framework to generate diverse and complex prompts, while structuring responses with explicit reasoning to aid models in accurately distinguishing safe from unsafe contexts. FalseReject includes training datasets tailored for both standard instruction-tuned models and reasoning-oriented models, as well as a human-annotated benchmark test set. Our extensive benchmarking on 29 state-of-the-art (SOTA) LLMs reveals persistent over-refusal challenges. Empirical results demonstrate that supervised finetuning with FalseReject substantially reduces unnecessary refusals without compromising overall safety or general language capabilities.
Forward citations
Cited by 7 Pith papers
-
IatroBench: Pre-Registered Evidence of Iatrogenic Harm from AI Safety Measures
Frontier models systematically withhold clinically necessary guidance from laypeople that they provide to physicians on identical facts, and LLM judges fail to detect that omission harm.
-
When Safety Becomes a Vulnerability: Exploiting LLM Alignment Homogeneity for Transferable Blocking in RAG
TabooRAG crafts a single adversarial document on a surrogate model that transfers across black-box RAG systems, causing many modern LLMs to refuse benign queries and reaching 96% ASR on GPT-5.2/HotpotQA.
-
BioTIER: A Refusal Benchmark for Targeted Biological Risk Mitigation
BioTIER, a 542-prompt benchmark with three risk tiers, shows frontier AI models differ by 90 percentage points in refusing dangerous biological queries, with top refusers over-refusing benign topics at the boundary.
-
From Refusal Geometry to Safety Geometry: Harmfulness--Refusal Coupling under Dynamic Adversarial Fine-Tuning
R2D2 fine-tuning of Mistral-7B transitions from high harmfulness–refusal coupling with collapsed utility to lower coupling with partial utility recovery and reopened jailbreaks; low coupling alone is not safety.
-
Robust Critics: Defending LLMs Against Multi-Turn Attacks
Critic-weighted sampling over inferred user intents improves multi-turn LLM defense success while preserving helpfulness, with an expected-Q improvement guarantee and transfer to frontier models.
-
The Personalization Trap: How User Memory Alters Emotional Reasoning in LLMs
Adding user memory to LLMs degrades their emotional-intelligence test scores and systematically disadvantages marginalized user profiles.
-
OPIUM: Mitigating Steering Externalities and Over-Refusal via Dual Objective Latent Optimization
OPIUM optimizes steering vectors in activation space so LLMs keep their intended behavior while shedding safety externalities and over-refusal, improving safety–utility tradeoff on Qwen-2.5 and LLaMA-3.1.
Discussion (0). Continue with ORCID to comment.